| 27 May 2026
Advancing the Scenario-to-Climate Link with Spatial Emulators in the FASTMIP Pilot Experiment
Abstract. Creating internally consistent climate scenarios remains a challenge in climate research. While Earth System Models (ESMs) offer detailed regional outcomes able to resolve and discover physical dynamics and drivers, their computational cost prevents them from covering the full range of emissions and land-use scenarios produced by Integrated Assessment Models (IAMs). As a result, the modelling chain from IAMs through Simple Climate Models (SCMs) to ESMs informing climate change assessments remains limited in its full deployment to a handful of scenarios, leaving most IAM scenarios lacking their corresponding regional climate information and limiting our ability to assess their implications for local impacts. This modelling chain also remains largely unidirectional: Climate outcomes are seldom fed back to inform the underlying scenarios whose development therefore may miss important potential feedback between physical climate and socioeconomic futures. Here, we introduce FASTMIP (Fast Assessment for Scenario Trajectories Multi-emulator Intercomparison Project), a coordinated effort to rapidly translate socioeconomic scenarios into spatial climate outcomes using emulators. In this FASTMIP pilot experiment, we apply three spatial emulators (MESMER, PRIME, and STITCHES) to a shared set of scenarios that includes both widely used CMIP6 pathways and scenarios without dedicated ESM simulations. The results show how spatial emulation could provide information on climate feedback to constrain scenario assumptions, extend scenario-to-climate coverage, support regionally differentiated assessments, and therefore reveal current limitations in the climate projection pipeline that limit a more comprehensive linkage and evaluation of scenario-to-climate outcomes. We show that for these 3 emulators there is consistency in the results, particularly for the center of the distribution. Addressing the limitations identified here at the scenario-to-climate interface could enable spatial emulators to systematically deliver rapid regional insights and support more consistent climate scenario development in future coordinated IAM-ESM modelling exercises.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Earth System Dynamics.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
This paper is both interesting an unsatisfying. Interesting in the angles that it brings, and unsatisfying because it is trying to do multiple things and do not do those well enough. Here are my specific points
Minor points;
Line 29: the fact that all three distributions agree on the center of distribution is probably of limited interested since the main goal of the discussed downscaling is to provide better extreme values. Additional points should be made.
Line 78: back to my main point #1, why isn’t there a tighter connection with the ScenarioMIP effort?
Section 3.1: why focus on fire-weather? This is already a very difficult process to represent on the ESM side, why is it thought that this approach would be representing those processes better?
Lines 166-168: how would that happen?
Lines 318-320: this is most likely just a reflection of the model selection, is it not?
Lines 347-349: I am not sure it demonstrates it. I think it shows there is potential, but I would want to see how GCAM/REMIND (since those are explicitly discussed in the paper, Figure 5) are planning to include this in their projections